AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Do Vision-Language Models Leak What They Learn? Adaptive Token-Weighted Model Inversion Attacks

Ngoc-Bao Nguyen, Sy-Tuyen Ho, Koh Jun Hao, Ngai-Man Cheung

CVPR 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Vision-language model robustness · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
aggregate losses over the whole output sequence instead of per-token updates (SMI vs TMI) (ablation)model inversion on LLaVA-v1.6-7B · FaceScrubAttAccM: 44.34% -> 59.17% (LLOM)Improved
add adaptive token weighting by visual attention (SMI-AW vs SMI) (ablation)SMI · FaceScrubAttAccM: 59.17% -> 61.01% (LLOM)Improved
add adaptive token weighting by visual attention (SMI-AW vs SMI) (ablation)SMI · FaceScrubAttAccM: 40.97% -> 41.16% (LCE)Improved
add adaptive token weighting by visual attention (SMI-AW vs SMI) (ablation)SMI · FaceScrubAttAccD Top1: 33.47% -> 37.62% (LLOM)Improved
add adaptive token weighting by visual attention (SMI-AW vs SMI) (ablation)SMI · CelebAAttAccD Top1: 38.30% -> 45.25%Improved
add adaptive token weighting by visual attention (SMI-AW vs SMI) (ablation)SMI · StanfordDogsAttAccM: 75.94% -> 78.13%Improved
add adaptive token weighting by visual attention (SMI-AW vs SMI) (ablation)SMI · StanfordDogsdelta_eval: 76.98 -> 81.66Hurt
update latent K times per token before moving on (TMI-C vs TMI) (ablation)TMI · FaceScrubAttAccM: 44.34% -> 31.16% (LLOM)Hurt
use max-margin loss instead of cross-entropy in SMI (ablation)SMI · FaceScrubAttAccM: 40.97% -> 55.52%Improved
use logit-maximization loss instead of cross-entropy in SMI-AW (ablation)SMI-AW · FaceScrubAttAccM: 41.16% -> 61.01%Improved
switch target VLM from LLaVA-v1.6-7B to Qwen2.5-VL-7BSMI-AW (LLOM) · FaceScrubAttAccM: 61.01% -> 32.03%Hurt
evaluate with human raters instead of MLLM judgeSMI-AW (LLOM) · FaceScrubattack accuracy: 61.01% (AttAccM) vs 56.93% (AttAccH)Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy